Announcement

Collapse
No announcement yet.

Issues in PWM circuit with LC filter to drive Peltier

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Issues in PWM circuit with LC filter to drive Peltier

    Hi,

    I've been trying to build a Peltier circuit that I can control with arduino and create cooling ramps. I've never built a circuit before and only have a basic knowledge of electronics.

    I've been researching how to do this and attempted to build it. My circuit is basically a PWM signal sent to a motor driver with an LC filter to smoothen the signal. Since very low temperatures are required, I added another Peltier on top of the first one. The top Peltier needs to be driven at a lower voltage to limit the amount of heat produced and effectively lower the temperature. I used a buck converter to do that (probably not the proper way to do this). With this setup, I can reach -53•C (hot side cooled by a waterbath at -15C) However, there are issues:


    - For the capacitors, I used 2 electrolytic of 470uF each. At low PWM duty cycle, they burn within seconds. The legs glow and come apart. That's how hot they get. I read about the ESR and the effects of frequency. I played around with the PWM frequency, from below 1kHz to 40kHz, but that didnt solve the issues. I tried several smaller uF and larger uF values without success. Then I read about ceramics capacitors. My local store only had 1 uF. I put a few of these in parallel but now the buck converter doesn't work properly, probably because a few uFs is not enough to smoothen the signal into a DC current for the buck converter. The ceramic capacitors get to about 50C at low duty cycles. I worry that larger ceramic capacitors will get hotter.

    - I got the parts from my local electronics store. They sold me an inductor that they said can handle high currents but couldn't tell me the inductance. It's about 7cm wide and long.

    - Currently, the PWM frequency is 31372.55 Hz.

    - Without being sure what causes the issue, I was going to get another inductor with a known inductance and several low ESR ceramic capacitors of 100 uF.

    - The current circuit is being supplied 12v from a 12v 50a power supply. To drive the Peltiers to their full potential, I will be getting a 24v 1000w power supply and a big heatsink. I'll need to keep the current below 25A to avoid killing my motor driver.

    The flyback diodes at 15A. There are 2 in parallel.


    The circuit is probably not drawn and built properly. Again, this is my first time doing this.



    The peltiers: https://cdn.badcaps-static.com/pdfs/...d4c890b119.pdf

    Motor driver: https://www.pololu.com/product/1455

    Buck converter (ideally I'd find an alternative to using a buck converter): https://www.creatroninc.com/product/...-with-display/

    Please let me know if I forgot to mention anything. Thank you for your help!
    Attached Files
    Last edited by vrasp; 10-27-2022, 08:20 PM.

    #2
    Re: Issues in PWM circuit with LC filter to drive Peltier

    You're using your flyback diodes wrong. Lose them and you'll save your capacitors.

    Comment


      #3
      Re: Issues in PWM circuit with LC filter to drive Peltier

      Thanks eccerr0r!

      I thought the flyback diodes were to allow voltage spike from the inductor to have somewhere to go when I cut the supply. Can you explain how am I using them wrong?

      Comment


        #4
        Re: Issues in PWM circuit with LC filter to drive Peltier

        If you step back a bit and look at what you made, you made a buck converter and the diode doesn't belong there... Like all buck converters, the inductor and capacitor are actually just filtering the output.

        Comment


          #5
          Re: Issues in PWM circuit with LC filter to drive Peltier

          Thank you! You saved me a ton of time. I removed the diodes and was able to run my cooling ramp.

          I am running into another issue, although a minor one. I'd like for the temperature to decrease at a constant rate. It is constant enough when I use one Peltier. However, with two, I get a rapid decrease of a few degrees as soon as the buck converter receives enough voltage to turn on the 2nd Peltier. So the cooling ramp starts at a constant rate, then a sudden drop of a few degrees, then a constant rate again.

          The only way I know of to maintain the rate constant is by having another PWM and LC filter connected to the 2nd Peltier which would start when the first PWM reaches 100% duty cycle. I then have complete control over the 2nd peltier cooling ramp.

          This is quite simple, However, if there is another way, maybe one where I don't have to have 2 PWM and filters, I'd love to know about it.

          Appreciated!

          Comment

          Working...
          X